The objective of this study was to examine patient history as well as clinical and laboratory features associated with true bacteremia versus false bacteremia in previously healthy febrile children ages 0 to 36 months in the era of polyvalent conjugate pneumococcal immunization.Methods
Using retrospective chart review, we examined history, physical examination, and laboratory characteristics associated with true and false bacteremia. We included subjects under 3 years old, with a positive blood culture obtained in the emergency department or clinic from July 2011 to July 2013, and fever defined as a temperature of greater than or equal to 100.4°F by history or examination. We excluded those with a previously known underlying disease process that could increase the risk for positive blood culture, for example, immunodeficiency, cancer, cystic fibrosis, or significant skin disorders such as severe eczema, as well as patients with any indwelling central line, shunt, or other implanted device, or recent hospitalization for a febrile illness.Results
Thirty subjects (24%) had true bacteremia, and 97 (76%) had false bacteremia. In the multivariable logistic regression analysis, exposure to sick contacts (odds ratio, 0.1; 95% confidence interval, 0.01–0.6; P = 0.01) and increased hours to positive blood culture (odds ratio, 0.8; 95% confidence interval, 0.8–0.9; P < 0.001) remained significant factors associated with false bacteremia. Age, maximum reported temperature, and reported days of fever were not associated with true bacteremia.Conclusions
We found that previously healthy children with true bacteremia are more likely to grow bacteria faster on blood culture and lack exposure to sick contacts than children with false bacteremia.